View in EDS

TS2Lang: A Co-occurrence pattern-driven translation mechanism for zero-shot time series forecasting with LLMs.

Saved in:
Bibliographic Details
Title: TS2Lang: A Co-occurrence pattern-driven translation mechanism for zero-shot time series forecasting with LLMs.
Authors: Xun, Haoran1 (AUTHOR) xhaoran@bupt.edu.cn, Zhou, Wen'an1 (AUTHOR) zhouwa@bupt.edu.cn, Tao, Liwen1 (AUTHOR) taoliwen@bupt.edu.cn, Zhong, Yingyi1 (AUTHOR) zyy2018@bupt.edu.cn
Source: Journal of Intelligent Information Systems. Apr2026, Vol. 64 Issue 2, p597-620. 24p.
Subjects: Sequential pattern mining, Vector quantization, Machine learning, Forecasting, Software frameworks, Language models, Autoregressive models
Abstract: With the rapid advancement of large language models (LLMs) in natural language processing and multimodal tasks, their potential in time series forecasting has attracted increasing attention. However, the representational differences between time series and text limit the effectiveness of transferring LLMs to temporal tasks. Existing approaches attempt to bridge this gap by learning discrete representations of time series, but these representations constitute a "foreign language" to LLMs, requiring additional continual pre-training and thus struggling to adapt to low-resource settings. To address this, we propose TS2Lang, a plug-and-play framework for time series forecasting with LLMs. TS2Lang first learns a general discrete representation of time series based on VQ-VAE and incorporates frequent sequence mining (FSM) to extract high-frequency patterns, effectively "translating" the time series into a form directly interpretable by LLMs. The resulting token sequences are then fed into a pre-trained LLM for autoregressive prediction, requiring no fine-tuning and supporting flexible replacement of the language model. In zero-shot settings, TS2Lang achieves the best average performance using far fewer trainable parameters and data than the strongest baseline, TimesFM, with MAE and MSE improved by up to approximately 13.4% and 7.9%, respectively. Moreover, adaptation with a small amount of target-domain data can further enhance performance, demonstrating the method's effectiveness and practical utility. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Intelligent Information Systems is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 193495186
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: TS2Lang: A Co-occurrence pattern-driven translation mechanism for zero-shot time series forecasting with LLMs.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Xun%2C+Haoran%22">Xun, Haoran</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> xhaoran@bupt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhou%2C+Wen'an%22">Zhou, Wen'an</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> zhouwa@bupt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Tao%2C+Liwen%22">Tao, Liwen</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> taoliwen@bupt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhong%2C+Yingyi%22">Zhong, Yingyi</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> zyy2018@bupt.edu.cn</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Journal+of+Intelligent+Information+Systems%22">Journal of Intelligent Information Systems</searchLink>. Apr2026, Vol. 64 Issue 2, p597-620. 24p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Sequential+pattern+mining%22">Sequential pattern mining</searchLink><br /><searchLink fieldCode="DE" term="%22Vector+quantization%22">Vector quantization</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Forecasting%22">Forecasting</searchLink><br /><searchLink fieldCode="DE" term="%22Software+frameworks%22">Software frameworks</searchLink><br /><searchLink fieldCode="DE" term="%22Language+models%22">Language models</searchLink><br /><searchLink fieldCode="DE" term="%22Autoregressive+models%22">Autoregressive models</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: With the rapid advancement of large language models (LLMs) in natural language processing and multimodal tasks, their potential in time series forecasting has attracted increasing attention. However, the representational differences between time series and text limit the effectiveness of transferring LLMs to temporal tasks. Existing approaches attempt to bridge this gap by learning discrete representations of time series, but these representations constitute a "foreign language" to LLMs, requiring additional continual pre-training and thus struggling to adapt to low-resource settings. To address this, we propose TS2Lang, a plug-and-play framework for time series forecasting with LLMs. TS2Lang first learns a general discrete representation of time series based on VQ-VAE and incorporates frequent sequence mining (FSM) to extract high-frequency patterns, effectively "translating" the time series into a form directly interpretable by LLMs. The resulting token sequences are then fed into a pre-trained LLM for autoregressive prediction, requiring no fine-tuning and supporting flexible replacement of the language model. In zero-shot settings, TS2Lang achieves the best average performance using far fewer trainable parameters and data than the strongest baseline, TimesFM, with MAE and MSE improved by up to approximately 13.4% and 7.9%, respectively. Moreover, adaptation with a small amount of target-domain data can further enhance performance, demonstrating the method's effectiveness and practical utility. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Journal of Intelligent Information Systems is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=193495186
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s10844-025-01015-6
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 24
        StartPage: 597
    Subjects:
      – SubjectFull: Sequential pattern mining
        Type: general
      – SubjectFull: Vector quantization
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Forecasting
        Type: general
      – SubjectFull: Software frameworks
        Type: general
      – SubjectFull: Language models
        Type: general
      – SubjectFull: Autoregressive models
        Type: general
    Titles:
      – TitleFull: TS2Lang: A Co-occurrence pattern-driven translation mechanism for zero-shot time series forecasting with LLMs.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Xun, Haoran
      – PersonEntity:
          Name:
            NameFull: Zhou, Wen'an
      – PersonEntity:
          Name:
            NameFull: Tao, Liwen
      – PersonEntity:
          Name:
            NameFull: Zhong, Yingyi
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 04
              Text: Apr2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 09259902
          Numbering:
            – Type: volume
              Value: 64
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: Journal of Intelligent Information Systems
              Type: main
ResultId 1